Electronic Supplementary Information (ESI):

MnO2 nanolayers on highly conductive TiO0.54N0.46

nanotubes for supercapacitor electrodes with high power

density and cyclic stability

Zhiqiang Wang, Zhaosheng Li*, Jianyong Feng, Shicheng Yan, Wenjun Luo, Jianguo

Liu, Tao Yu, and Zhigang Zou*

National Laboratory of Solid State Microstructures, Ecomaterials and Renewable Energy

Research Center (ERERC), Department of Physics, Nanjing University, 22 Hankou Road,

Nanjing 210093, People’s Republic of China.

Experimental and calculation details

The area capacitance (in mF cm−2) of the electrodes from the cyclic voltammetry were

calculated according to the following equation:

, (1)Idt

CE S

where I is the discharge current (in mA) of the discharge process, ΔE is the potential

window (0.8 V) and S is the surface area of the electrode (in cm2). The integration

was conducted over one complete discharge process.

The mass capacitance (in F g−1) of the electrode was calculated from the

galvanostatic charge-discharge curve using to the following equation:

, (2)I tCE m

where I is the discharge current (in A) of the discharge process, Δt is the time required

for the voltage to vary from 0.8 V to 0 V, ΔE is the potential window (0.8 V), and m is

the mass of MnO2 on the electrode.

Electronic Supplementary Material (ESI) for Physical Chemistry Chemical Physics.This journal is © The Royal Society of Chemistry 2014

For the two electrode system and the devices, the total capacitance (C), energy

densities (E) and power densities (P) were calculated by using the following equations:

𝐶 = 𝐼∆𝑡∆𝑉 ,

𝐸 = 12𝐶(∆𝑉)2 ,

𝑃 = (∆𝑉)2

4𝑅𝑀

where I, Δt, ΔV, R, and M are the discharge current density, the discharge time,

the voltage window after the deduction of the IR drop in discharge, the internal

resistance from IR drop, and the total mass of MnO2 in both electrodes, respectively.

The internal resistances (R) were determined from the IR drop, i.e. voltage drop

ΔVdrop at the beginning of the galvanostatic discharge curves:

𝑅 =∆𝑉𝑑𝑟𝑜𝑝

2𝐼

Figure S1. Current and voltage curves of the two electrode system in a classic

anodization process.

Figure S2. XRD patterns of the titania nanotubes after they were nitrificated at 600

°C.

Figure. S3. Dominant XRD peak (200) positions of the TiO1-xNx nanotubes after they

were nitrificated at 650, 700, 800 and 900 °C.

As the nitrification temperature increased, the dominant XRD peak (200)

sharpened and the center of the peak shifted to lower 2 theta angles, indicating that the

N concentration in TiO1-xNx increased with the nitrification temperature. XRD

patterns confirmed that the amorphous titanate (XRD are not shown here) transformed

to cubic TiOxN1-x when the nitrification temperature was above 650 °C.

Figure S4. Effect of the nitrification temperature on the component and capacitance

of the TiO1-xNx nanotube arrays.

Figure S5. Effect of various aqueous electrolytes (LiCl, NaCl, KCl, RbCl and CsCl)

on the performance of TiO1-xNx nanotube arrays. A three electrode configuration with

a scan rate of 0.1 V/s was used. The concentration of the electrolyte was 1.0 mol L−1

for all samples.

Figure S6. Galvanostatic charge/discharge curves for a three electrode configuration

in a 1 mol L−1 KCl aqueous solution for (a) TiO0.54N0.46 nanotube arrays and (b)

MnO2 on TiO0.54N0.46 nanotube arrays (MnO2 was deposited for 50 s) at various

current densities.

Figure S7. Specific capacitances of TiOxN1-x nanotube array electrodes calculated

using cyclic voltammetray as a function of the scan rate (three electrode configuration,

in a 1 mol L−1 KCl aqueous solution).

Figure S8. (a) Cyclic voltammetry (0.1 V s−1) of titania nanotube arrays annealed in

various gases. (b) Electrochemical impedance spectroscopy (EIS), 0.1 – 10,000 Hz.

For comparison air, oxygen, a hydrogen/nitrogen (1:9) gas mixture and argon

were used as the atmosphere in the titania nanotube arrays annealing procedure

instead of NH3 gas. The impedances of these samples were larger than the samples

annealed in NH3 (Fig. 4f). The cyclic voltammetry curves deviated from a rectangular

shape for a much lower charge/discharge current and capacitance, compared with

those annealed in NH3.

Figure S9. SEM and TEM images of the samples. (a) A titanate nanotube array; (b),

(c) and (d) are TiO1-xNx nanotube arrays nitrificated at 700 °C. (e) and (f) are images

of a MnO2/TiO0.54N0.46 interface. The inset in (c) shows the selected area electron

diffraction (SEAD) pattern.

Fig. S9a showed that well aligned titanate nanotube arrays with a tube diameter

of ~150 nm were formed on flexible titanium foil after anodization and the titanate

membranes were connected through regular nanotubes. This enhanced the mechanical

strength of the nanotube arrays, inhibiting exfoliation.

Fig. S9b, c and d revealed that TiO0.54N0.46 formed nanotube arrays after

nitrification at 700 °C. After the nitrification process at temperatures of 650, 700 and

800 °C, the cubic TiO1-xNx nanotube arrays had similar morphologies to the titanate

nanotube arrays, excluding the nanopores on the walls of the nanotubes. Some

nanopores in the nanotube walls were observed in the TiO0.54N0.46 samples, indicating

that TiO0.54N0.46 exhibited a hierarchic nanostructure.

Fig. S9c and d showed that many nanopores existed in the walls of the

TiO0.54N0.46 nanotubes. The inset in Fig. S9c shows the selected area electron

diffraction image of a TiO0.54N0.46 nanotube array. This suggested that the TiO0.54N0.46

nanotubes were highly crystalline.

Fig. S9e and f showed that MnO2 was loaded onto the TiO0.54N0.46 nanotubes.

The surface of the TiO0.54N0.46 was covered in MnO2 with a thickness of 5 – 10 nm for

the 100 s deposited samples, which favors improving the conductance.

Figure S10. The cross-sectional view of the MnO2/TiO0.54N0.46/Ti.

Figure S11. Effect of the MnO2 loading time on the current density-specific

capacitance relationship of MnO2 on TiO0.54N0.46 nanotube arrays (in a three electrode

configuration in a 1 mol L−1 KCl aqueous solution).

Figure S12. Effect of the MnO2 loading time on the IR drop for MnO2 on TiO0.54N0.46

nanotube arrays at a current densities of (a) 1000A/g and (b) 2000A/g for a three

electrode configuration in a 1 mol L−1 KCl aqueous solution.

Figure S13. Comparison of the capacitance (in a 1 mol L−1 KCl aqueous solution) of

TiO0.54N0.46 nanotube arrays with the MnO2-loaded TiO0.54N0.46 nanotube arrays for

100 s and 400 s MnO2 deposited samples at various current densities.

Table S1. Current density in galvanostatic test in Figure S13 (mA cm−2)

TiO0.54N0.46 100s MnO2-deposited

TiO0.54N0.46

400s MnO2-deposited

TiO0.54N0.46

0.131 0.131

0.262 0.262

0.521 0.521

0.655 0.655

1.042 1.042

1.31 1.31

2.605 2.605

2.62 2.62

5.21 5.21

6.55 6.55

10.42 10.42

13.1 13.1

26.05 26.05

26.2 26.2

52.1 52.1

In Figure S13, the specific area capacitance of the TiO0.54N0.46 nanotubes was

measured under the same area current density as the MnO2-deposited TiO0.54N0.46

samples. Note that all of the pseudo capacitive materials had an EDLC portion in their

capacitance. The capacitance of the TiO0.54N0.46 nanotubes was regarded mainly

contributed from the electrochemical double layer capacitance in the mild electrolyte

of 1 mol L−1 KCl aqueous solution. From Figure S13, it was concluded that the EDLC

only took up a very small portion of the total capacitance, and the capacitance of the

MnO2/TiO0.54N0.46 electrode mainly comes from the pseudo capacitance of MnO2.

The mass of the TiO0.54N0.46 nanotubes were causiously weighed with a micro

balance (METTLER TOLEDO, MX 5) after scratch them from the substrate. The

mass of TiO0.54N0.46 nanotubes was determined to be 0.12 mg cm−2.

Table S2. Equivalent circuit parameters obtained from the fitting results

(corresponding to Fig. 4f). This version of the Warburg element was terminated to be

an open circuit.

Loading time

of MnO2

R1 R2 C1 W1-R W1-T W1-P

0 s 1.96 0.512 3.76×10−5 1.83 7.19×10−3 0.476

50 s 2.83 0.878 4.89×10−5 1.43 1.36×10−2 0.464

100 s 2.90 1.15 5.52×10−5 1.20 1.80×10−2 0.461

200 s 3.07 1.28 3.99×10−5 1.43 3.37×10−2 0.468

400 s 3.41 1.77 3.17×10−5 2.63 0.116 0.474

Figure S14. Effect of the folding angle on the capacitance (in a 1 mol L−1 KCl

aqueous solution) of MnO2 on TiO0.54N0.46 nanotube arrays (r = 5 mm).

Figure S15. Effect of the folding time on the capacitance (in a 1 mol L−1 KCl aqueous

solution) of MnO2 on TiO0.54N0.46 nanotube arrays. (r = 5mm, θ ranged from −180 ° to

180 °)

Figure S16. Capacity retention of MnO2/TiO0.54N0.46 supercapacitor devices (with a 1

mol L−1 KCl aqueous solution). A capacity retention of 97.5% was recorded after

10,000 galvanostatic cycles.

Figure S17. Energy and power densities (Ragone plot) of MnO2 in MnO2/TiO0.54N0.46

supercapacitor devices.

Figure S18. (a) The schematic diagram of the supercapacitor device. (b) A red light

emitting diode (LED) powered by two MnO2/TiO0.54N0.46 supercapacitor devices

connected in series.

The supercapacitor devices were assembled by two pieces of 100 s deposited

MnO2/TiO0.54N0.46 electrodes with a separator (glass fiber paper with pore diameter of

0.26 μm) sandwiched in between, and with 1 mol L−1 KCl aqueous solution as

electrolyte. Prior to the electrolyte infiltration, the electrodes and the separator were

assembled into a rectangular plastic bag (LDPE, low density polyethylene) with sides

sealed by 300 °C heated air except the inlet, set aside for the electrolyte injection.

After electrolyte injected and infiltration into the electrode and separator, the inlet was

sealed by the 300 °C heated air as before.

Figure S19. (a) Photo showing the bending of the MnO2/TiO0.54N0.46 electrode in 180

°. (b) Photo showing the bending of the MnO2/TiO0.54N0.46 supercapacitor devices in

180 °.

Figure S20. Energy and power densities of MnO2 compared with that of solely

TiO0.54N0.46 (Ragone plot, measured in a two-electrode configuration in the discharge

current of 100 A g−1).

Table S3. Comparison of the power density and energy density of the MnO2 and

TiO0.54N0.46（the voltage window, 0-0.8 V; 1 mol L-1 KCl aqueous）

Active materials Power density (kW kg−1) Energy density (Wh kg−1)

MnO2 on TiO0.54N0.46 620.2 9.8

TiO0.54N0.46 708.0 2.1

The energy and power densities of MnO2 compared with that of solely

TiO0.54N0.46 nanotubes array have been shown in the Figure S20 and Table S3.

Compared to the solely TiO0.54N0.46 nanotubes array, the MnO2 on the TiO0.54N0.46

show a relatively higher energy density, because the faradaic reaction of the MnO2.

While due to the low reaction rate of the faradaic reaction, the discharge rate are

limited, then the MnO2 show a relatively lower power density compared to the solely

TiO0.54N0.46 nanotubes array that accommodate charges in double layer mechanism.